Confounding Influences on Police Detection of Suspiciousness
Richard R. JohnsonJournal of Criminal JusticeVol. 34 No. 4, 2006pp. 435-442,
Johnson (2006) analyzes police-citizen encounters to determine whether citizen race or emotional agitation level influence the frequency with which the citizens display nonverbal behaviors generally considered “suspicious”by criminal justice professionals. In particular, Johnson attempts to determine whether the frequency of speech disruptions, smiles, gaze aversions, and hand gestures differ across racial groups and those who do/do not exhibit signs of emotional agitation during face-to-face officer-citizen interactions where the citizen is not suspected of a crime. Johnson cites findings of experiments conducted in controlled laboratory environments that suggest the display of suspicious behavioral cues are influenced by both stress and cultural differences. The author expands upon previous research by examining citizen behaviors in real life scenarios.
Video footage of interactions between citizens and police officers from the reality television show “Cops”, aired during the years 2002 and 2003,was analyzed. Cases were selected from scenes in which a non-suspect citizen engaged in verbal interaction with an officer and was visible for at least sixty seconds from the front and from the waist up. From the 223 interactions that met these criteria, a stratified sample of 120 cases was randomly selected based on race (African Americans, Caucasians, and Hispanics) and level of agitation(agitated and not agitated). For each case, the number of citizen smiles, speech disruptions (“ahs,” “ums,” or stutters), seconds of continuous eye contact, and hand gestures were counted. These numbers were then converted into a rate per minute for each nonverbal behavior.
While no significant relationships were found between speech disturbances or eye contact and emotional agitation, Johnson found that citizens who were visibly emotionally agitated were significantly less likely to smile and significantly more likely to use hand gestures than those who did not display agitation. Initial multivariate two-way analysis of variance (ANOVA) tests also revealed significant differences in the frequency of all four nonverbal behaviors displayed across racial groups, even while controlling for level of emotional agitation. Johnson used a series of t-tests to compare the behaviors of African-American and Hispanics to Caucasians. The author found that during police contacts:
African-Americans were significantly more likely to smile while Hispanics were significantly less likely to smile than Caucasians.
African-Americans were significantly less likely to display speech disruptions while Hispanics were significantly more likely to do so than Caucasians.
Both African-Americans and Hispanics were significantly less likely to maintain eye contact with officers than Caucasians.
Both African-Americans and Hispanics were significantly more likely to use hand gestures than Caucasians.
Johnson suggests that many of these significant differences in nonverbal cues may cause police officers to identify the behavior of African-Americans and Hispanics as “suspicious” more often, even when they are not engaging in misconduct. Johnson suggests that these behavioral differences across racial groups may help to explain recent findings that African-Americans and Hispanics are significantly more likely to be searched during traffic stops even after controlling for involvement in crime. The author recommends that police agencies reconsider the use of training curriculum that teaches police officers to rely on smiles, speech disruptions, avoidance of eye contact, and hand gestures to detect suspicious behaviors until further research can be conducted.
Tamara D. Madensen University of Nevada, Las Vegas, USA
